Khairul Umam Syaliman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Khairul Umam Syaliman has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Adli Abdillah Nababan
Bina Nusantara University
1 shared publication Aldi Abdillah Nababan
Universitas Sumatera Utara
1 shared publication- Edi Susilo
Universitas Riau
1 shared publication - Erna Budhiarti Nababan
Universitas Sumatera Utara
1 shared publication M. Zulfahmi
Universitas Sumatera Utara
1 shared publication- Opim Salim Sitompul
Universitas Sumatera Utara
1 shared publication
Latest Publications
- Open Access
- 22 October 2026
- Published by Private Company Technology Center in Eastern-European Journal of Enterprise Technologies
- Vol. 5 (4 (143)) , 35-45
Identification of dataset- and classifier-dependent performance patterns in latin hypercube sampling with nearest-neighbor rebalancing for imbalanced classification
The object of the study is the dataset- and classifier-dependent behavior of a frozen Latin hypercube sampling and nearest-neighbor rebalancing procedure for imbalanced tabular classification. The problem addressed was the absence of practical evidence showing when rebalancing improves, leaves unchanged, or degrades performance across datasets and classifier families. The procedure was evaluated on 20 tabular datasets using five deterministic seeds, duplicate-aware cross-validation, eight sampling conditions, and four classifiers. Relative to synthetic minority over-sampling, it increased dataset-balanced macro-averaged F1 score from 0.7717 to 0.7761, produced 13 dataset-level wins and 7 losses, and yielded a Wilcoxon p-value of 0.0441 with a paired bootstrap 95% confidence interval of [0.00104, 0.00788]. The mean dataset-level difference was +0.004389. The largest classifier-specific gain in the dataset-balanced summaries occurred with distance-weighted seven-nearest neighbors (+0.011944); gains for extreme gradient boosting, random forest, and multilayer perceptron were small. Six dataset characteristics showed no significant monotonic association with gains after Holm correction. This does not establish causal effects of local geometry or classifier induction. The distinctive result is a reproducible classification of benefit, neutral, and loss regimes supported by dataset-level inference, classifier contrasts, minority-class outcomes, and runtime analysis. Mean total runtime was 2.89 times that of the SMOTE comparator. The findings guide comparative validation when local component preservation, controlled synthetic generation, and neighborhood-compatible classification are important
- 22 October 2026
- Published by Private Company Technology Center in Eastern-European Journal of Enterprise Technologies
- Vol. 5 (4 (143)) , 35-45
Open AccessIdentification of dataset- and classifier-dependent performance patterns in latin hypercube sampling with nearest-neighbor rebalancing for imbalanced classification
The object of the study is the dataset- and classifier-dependent behavior of a frozen Latin hypercube sampling and nearest-neighbor rebalancing procedure for imbalanced tabular classification. The problem addressed was the absence of practical evidence showing when rebalancing improves, leaves unchanged, or degrades performance across datasets and classifier families. The procedure was evaluated on 20 tabular datasets using five deterministic seeds, duplicate-aware cross-validation, eight sampling conditions, and four classifiers. Relative to synthetic minority over-sampling, it increased dataset-balanced macro-averaged F1 score from 0.7717 to 0.7761, produced 13 dataset-level wins and 7 losses, and yielded a Wilcoxon p-value of 0.0441 with a paired bootstrap 95% confidence interval of [0.00104, 0.00788]. The mean dataset-level difference was +0.004389. The largest classifier-specific gain in the dataset-balanced summaries occurred with distance-weighted seven-nearest neighbors (+0.011944); gains for extreme gradient boosting, random forest, and multilayer perceptron were small. Six dataset characteristics showed no significant monotonic association with gains after Holm correction. This does not establish causal effects of local geometry or classifier induction. The distinctive result is a reproducible classification of benefit, neutral, and loss regimes supported by dataset-level inference, classifier contrasts, minority-class outcomes, and runtime analysis. Mean total runtime was 2.89 times that of the SMOTE comparator. The findings guide comparative validation when local component preservation, controlled synthetic generation, and neighborhood-compatible classification are important
- Open Access
- 16 March 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 978 (1) , 012047
Improving the accuracy of k-nearest neighbor using local mean based and distance weight
- 16 March 2018
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 978 (1) , 012047
Open AccessImproving the accuracy of k-nearest neighbor using local mean based and distance weight
- Open Access
- 3 September 2017
- Published by Universitas Islam Sumatera Utara in InfoTekJar (Jurnal Nasional Informatika dan Teknologi Jaringan)
- Vol. 2 (1) , 79-89
Perbandingan Rapid Centroid Estimation (RCE) — K Nearest Neighbor (K-NN) Dengan K Means — K Nearest Neighbor (K-NN)
Teknik Clustering terbukti dapat meningkatkan akurasi dalam melakukan klasifikasi, terutama pada algoritma K-Nearest Neighbor (K-NN). Setiap data dari setiap kelas akan membentuk K cluster yang kemudian nilai centroid akhir dari setiap cluster pada setiap kelas data tersebut akan dijadikan data acuan untuk melakukan proses klasifikasi menggunakan algoritma K-NN. Namun kendala dari banyaknya teknik clustering adalah biaya komputasi yang mahal, Rapid Centroid Estimation (RCE) dan K-Means termasuk kedalam teknik clustering dengan biaya komputasi yang murah. Untuk melihat manakah dari kedua algoritma ini (RCE dan K-Means) yang lebih baik memberikan peningkatan akurasi pada algoritma K-NN maka, pada penelitian ini akan mencoba untuk membandingkan kedua algoritma tersebut. Hasil dari penelitian ini adalah gabungan RCE—K-NN memberikan hasil akurasi yang lebih baik dari K-Means—K-NN pada data set iris dan wine. Namun dalam perubahan nilai akurasi RCE—K-NN lebih stabil hanya pada data set iris. Sedangkan pada data set wine, K-Means—K-NN terlihat mendapati perubahan akurasi yang lebih stabil dibandingkan RCE—K-NN.
- 3 September 2017
- Published by Universitas Islam Sumatera Utara in InfoTekJar (Jurnal Nasional Informatika dan Teknologi Jaringan)
- Vol. 2 (1) , 79-89
Open AccessPerbandingan Rapid Centroid Estimation (RCE) — K Nearest Neighbor (K-NN) Dengan K Means — K Nearest Neighbor (K-NN)
Teknik Clustering terbukti dapat meningkatkan akurasi dalam melakukan klasifikasi, terutama pada algoritma K-Nearest Neighbor (K-NN). Setiap data dari setiap kelas akan membentuk K cluster yang kemudian nilai centroid akhir dari setiap cluster pada setiap kelas data tersebut akan dijadikan data acuan untuk melakukan proses klasifikasi menggunakan algoritma K-NN. Namun kendala dari banyaknya teknik clustering adalah biaya komputasi yang mahal, Rapid Centroid Estimation (RCE) dan K-Means termasuk kedalam teknik clustering dengan biaya komputasi yang murah. Untuk melihat manakah dari kedua algoritma ini (RCE dan K-Means) yang lebih baik memberikan peningkatan akurasi pada algoritma K-NN maka, pada penelitian ini akan mencoba untuk membandingkan kedua algoritma tersebut. Hasil dari penelitian ini adalah gabungan RCE—K-NN memberikan hasil akurasi yang lebih baik dari K-Means—K-NN pada data set iris dan wine. Namun dalam perubahan nilai akurasi RCE—K-NN lebih stabil hanya pada data set iris. Sedangkan pada data set wine, K-Means—K-NN terlihat mendapati perubahan akurasi yang lebih stabil dibandingkan RCE—K-NN.